Gradient Penalty not Used in WGANGP

I noticed the calculation of wgangp [here](https://github.com/Ha0Tang/AttentionGAN/blob/68a478a944bb45d288d67f99fe110ddf087fd84d/models/networks.py#L274)
did not use [the gradient penalty](https://github.com/Ha0Tang/AttentionGAN/blob/68a478a944bb45d288d67f99fe110ddf087fd84d/models/networks.py#L282).
Is it just a code stub?